view sat/tools/web.py @ 3922:0ff265725489

plugin XEP-0447: handle attachment and download: - plugin XEP-0447 can now be used in message attachments and to retrieve an attachment - plugin attach: `attachment` being processed is added to `extra` so the handler can inspect it - plugin attach: `size` is added to attachment - plugin download: a whole attachment dict is now used in `download` and `file_download`/`file_download_complete`. `download_uri` can be used as a shortcut when just a URI is used. In addition to URI scheme handler, whole attachment handlers can now be registered with `register_download_handler` - plugin XEP-0363: `file_http_upload` `XEP-0363_upload_size` triggers have been renamed to `XEP-0363_upload_pre_slot` and is now using a dict with arguments, allowing for the size but also the filename to be modified, which is necessary for encryption (filename may be hidden from URL this way). - plugin XEP-0446: fix wrong element name - plugin XEP-0447: source handler can now be registered (`url-data` is registered by default) - plugin XEP-0447: source parsing has been put in a separated `parse_sources_elt` method, as it may be useful to do it independently (notably with XEP-0448) - plugin XEP-0447: parse received message and complete attachments when suitable - plugin XEP-0447: can now be used with message attachments - plugin XEP-0447: can now be used with attachments download - renamed `options` arguments to `extra` for consistency - some style change (progressive move from legacy camelCase to PEP8 snake_case) - some typing rel 379
author Goffi <goffi@goffi.org>
date Thu, 06 Oct 2022 16:02:05 +0200
parents 65bac82e4049
children 524856bd7b19
line wrap: on
line source

#!/usr/bin/env python3

# Libervia: an XMPP client
# Copyright (C) 2009-2021 Jérôme Poisson (goffi@goffi.org)

# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Affero General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.

# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
# GNU Affero General Public License for more details.

# You should have received a copy of the GNU Affero General Public License
# along with this program.  If not, see <http://www.gnu.org/licenses/>.

from typing import Optional, Union
from pathlib import Path
from io import BufferedIOBase

from OpenSSL import SSL
import treq
from treq.client import HTTPClient
from twisted.internet import reactor, ssl
from twisted.internet.interfaces import IOpenSSLClientConnectionCreator
from twisted.web import iweb
from twisted.web import client as http_client
from zope.interface import implementer

from sat.core import exceptions
from sat.core.log import getLogger


log = getLogger(__name__)


SSLError = SSL.Error


@implementer(IOpenSSLClientConnectionCreator)
class NoCheckConnectionCreator(object):
    def __init__(self, hostname, ctx):
        self._ctx = ctx

    def clientConnectionForTLS(self, tlsProtocol):
        context = self._ctx
        connection = SSL.Connection(context, None)
        connection.set_app_data(tlsProtocol)
        return connection


@implementer(iweb.IPolicyForHTTPS)
class NoCheckContextFactory:
    """Context factory which doesn't do TLS certificate check

    /!\\ it's obvisously a security flaw to use this class,
    and it should be used only with explicit agreement from the end used
    """

    def creatorForNetloc(self, hostname, port):
        log.warning(
            "TLS check disabled for {host} on port {port}".format(
                host=hostname, port=port
            )
        )
        certificateOptions = ssl.CertificateOptions(trustRoot=None)
        return NoCheckConnectionCreator(hostname, certificateOptions.getContext())


#: following treq doesn't check TLS, obviously it is unsecure and should not be used
#: without explicit warning
treq_client_no_ssl = HTTPClient(http_client.Agent(reactor, NoCheckContextFactory()))


async def downloadFile(
    url: str,
    dest: Union[str, Path, BufferedIOBase],
    max_size: Optional[int] = None
) -> None:
    """Helper method to download a file

    This is for internal download, for high level download with progression, use
    ``plugin_misc_download``.

    Inspired from
    https://treq.readthedocs.io/en/latest/howto.html#handling-streaming-responses

    @param dest: destination filename or file-like object
        of it's a file-like object, you'll have to close it yourself
    @param max_size: if set, an exceptions.DataError will be raised if the downloaded file
        is bigger that given value (in bytes).
    """
    if isinstance(dest, BufferedIOBase):
        f = dest
        must_close = False
    else:
        dest = Path(dest)
        f = dest.open("wb")
        must_close = True
    d = treq.get(url, unbuffered=True)
    written = 0

    def write(data: bytes):
        if max_size is not None:
            nonlocal written
            written += len(data)
            if written > max_size:
                raise exceptions.DataError(
                    "downloaded file is bigger than expected ({max_size})"
                )
        f.write(data)

    d.addCallback(treq.collect, f.write)
    try:
        await d
    except exceptions.DataError as e:
        log.warning("download cancelled due to file oversized")
        raise e
    except Exception as e:
        log.error(f"Can't write file {dest}: {e}")
        raise e
    finally:
        if must_close:
            f.close()